Artificial intelligence? But of course!
Dietikon, 13 July 2021
You can’t have parcels delivered right to your door AND protect the environment, can you? Yes, you can. That’s exactly why we plan the ‘Planzer Parcel’ routes with fast, self-learning algorithms.
Digital is the norm
Today, ordering items online and having them delivered directly to your door is seen as the norm – as standard a part of our digital lives as chatting with friends on WhatsApp or video conferencing with work colleagues.
As a logistics partner, if we drove around Switzerland in our delivery van delivering each ordered item individually, it would be a huge waste of limited resources such as fuel and of productivity. Not to mention logistically impossible. As a family-run business, we value sustainability and innovation – which is why we choose to rely on the intelligence of cutting-edge algorithms. And on the long-standing expertise of the Fraunhofer Institute for Material Flow and Logistics, IML.
How does it work?
A little IT background for starters: what is an algorithm? An algorithm is a series of instructions that are calculated and followed step by step in order to complete a task. In our case, the task is efficient route planning. The rules aren’t invented by some programmer. They are the result of a calculation that takes into account influencing factors, criteria and priorities.
Meeting a range of requirements
Together with the IML, we developed a self-learning algorithm and provided it with a range of influencing factors, criteria and priorities. These included orders with a same-day collection or delivery date, the number of available drivers, distances, stop times and much more. Using these specifications, our algorithm sets about its calculations the night before the delivery date and outputs the optimal routes. In addition to the above specifications, these routes have to take account of redeliveries, the time frames you have requested and – most important of all – ensure that we keep the number of kilometres we travel to a minimum.
Research for real-world applications
We have been working with the IML since 2017 to ensure that we make the most of the possibilities that new technologies offer the national transport sector. What began as a research project has turned into an intensive collaboration for applied research. We have long since moved on from pure mathematical problems. We are researching which algorithm offers the most efficient route planning for our growing volume of deliveries. There is no such thing as a single algorithm that will solve every problem. Today, we are already working with our second version and are about to introduce the third.
Never done learning
Since October 2020, we have also been working with the IML on a ‘learning system’. Our development team consists of specialists in technologies and methods of using artificial intelligence. The aim of this co-creation is to reinforce our route planning algorithm with a learning system. This will allow the algorithm to move beyond the specified criteria and use artificial intelligence and machine learning, further optimising route planning at ‘Planzer Parcel’.
Good to know
You need have no qualms about having your parcels delivered right to your door. Because now you know that we are both helping protect the climate thanks to the intelligence of advanced digital technologies.
